Prediction of Donor Heart Acceptance for Transplant and Its Clinical Implications: Results From The Donor Heart Study

Abstract

Background: Despite a shortage of potential donors for heart transplant in the United States, most potential donor hearts are discarded. We evaluated predictors of donor heart acceptance in the United States and applied machine learning methods to improve prediction.

Methods: We included a nationwide (2005-2020) cohort of potential heart donors in the United States (n=73 948) from the Scientific Registry of Transplant Recipients and a more recent (2015-2020) rigorously phenotyped cohort of potential donors from DHS (Donor Heart Study; n=4130). We identified predictors of acceptance for heart transplant in both cohorts using multivariate logistic regression, incorporating time-interaction terms to characterize their varying effects over time. We fit models predicting acceptance for transplant in a 50% training subset of DHS using logistic regression, least absolute shrinkage and selection operator, and random forest algorithms and compared their performance in the remaining 50% (test) of the subset.

Results: Predictors of donor heart acceptance were similar in the nationwide and DHS cohorts. Among these, older age (P value for time interaction, 0.0001) has become increasingly predictive of discard over time while other factors, including those related to drug use, infection, and mild cardiac diagnostic abnormalities, have become less influential (P value for time interaction, <0.05 for all). A random forest model (area under the curve, 0.908; accuracy, 0.831) outperformed other prediction algorithms in the test subset and was used as the basis of a novel web-based prediction tool.

Conclusions: Predictors of donor heart acceptance for transplantation have changed significantly over the last 2 decades, likely reflecting evolving evidence regarding their impact on posttransplant outcomes. Real-time prediction of donor heart acceptance, using our web-based tool, may improve efficiency during donor management and heart allocation.

Keywords: donor selection; heart transplantation; machine learning; random forest; tissue and organ procurement.

Figure 1:. Forest plots of adjusted associations between donor characteristics and acceptance for HT in (a) DHS and (b) nationwide cohorts and their change over time

Shown are odds ratios and 95% Cis (in brackets) from multivariable logistic regression models as detailed in Methods. Referent groups include Age < 35 years, normal coronary angiogram, normal LV function (LVEF ≥ 50%), male sex, anoxia or other cause of death, ABO type O, absence of LV hypertrophy, and troponin < 10 x ULN. Arrows (panel b) indicate the direction of all statistically significant time interactions. Abbreviations: CAD: coronary artery disease, DHS: Donor Heart Study, HT: heart transplantation, LV: left ventricular, NT pro-BNP: NT-pro B-type natriuretic peptide, PHS: Public Health Service

Figure 2:. Receiver operating characteristic curves (a) and calibration plot (b; Random Forest model only) for prediction models in test dataset

Abbreviations: CI: confidence interval; LASSO: least absolute shrinkage and selection operator

Figure 3:. Screenshot of web-based tool predicting heart acceptance for transplant (ToP-HAT)

Abbreviations: BNP: B-type natriuretic peptide, DHS: Donor Heart Study, LV: left ventricular, NT pro-BNP: NT-pro B-type natriuretic peptide, PHS: Public Health Service

Figure 4:. A potential scenario for implementation of ToP-HAT during donor evaluation

For many potential donors, their viability for use in HT is not clear at the early stages of donor management. Making this determination can require extensive cardiac workup including serial echocardiograms and/or a coronary angiogram; depending on local availability, the latter may require transfer to a different center. The resulting delay in organ recovery is resource-intensive, including augmented inotropic therapy and administration of thyroid hormone or corticosteroids, and can compromise the quality of other solid organs, particularly when the donor is hemodynamically unstable. The probability of acceptance for HT – estimated early on using ToP-HAT – can help guide the decision to pursue further cardiac workup. When this probability fails to meet some reasonable threshold, then deferring evaluation for potential HT may be warranted. Abbreviations: HT: heart transplantation, ToP-HAT: Tool Predicting Heart Acceptance for Transplantation, TTE: transthoracic echocardiogram